fbpx

First HANA performance results released

By Freya Purnell

SAP In-Memory Appliance (or SAP HANA) can handle 10,000 queries per hour against 1.3 terabytes of data, returning results within seconds, according to the appliance’s first official performance results, released by SAP and IBM last week.

The test scenarios were based on typical query workload requirements seen in SAP customers’ ERP reporting, with data found in sales delivery database tables.

Part of a joint project between IBM and SAP to deliver real-time analytics by optimising SAP in-memory computing technology and IBM’s 3850 series servers, the test used testing scenarios designed to measure the performance of the in-memory database processing analytical queries on operational business data. The  scenarios closely mirrored the complexity of enterprise data, in terms of data definition, data types and data distributions, without pre-aggregating the data or tuning the database.

The test was independently audited and verified by WinterCorp, and president and founder, Richard Winter said, “In a review of the test specification, test runs and outputs, I find that SAP HANA performance test is a valid determination of data warehouse query performance on in-memory test databases up to and including four terabytes of realistically generated test data. The test data values are closely patterned after real SAP data, and the test queries are similar to operational and analytical business intelligence queries that occur in practice.”

SAP and IBM used this approach in order to more accurately recreate the data that is stored in actual customer systems and the real-time requirements of business intelligence (BI) systems.

In the 1.3 terabyte test, SAP In-Memory Computing technology was packaged with an IBM X5 server containing 32 cores, 0.5 terabytes of memory and a RAID 5 disk system, with results demonstrating an excellent cost-to-performance ratio.

“By dramatically outperforming traditional disk-based systems on analytical queries, SAP HANA opens new business opportunities and makes real-time business possible for the first time,” said Jake Klein, vice president, Data and Analytic Engines, SAP. “The results of this performance test are one more important step for SAP and our hardware partners, demonstrating the significance of our in-memory computing technology to both the industry and our customers.”

IBM and SAP have cooperated closely to implement the advanced replication scenario for DB2-based customers. According to SAP, DB2 is HANA-ready and can efficiently replicate data into SAP HANA in near real-time using the Sybase Replication Server.

Share this post

submit to reddit
scroll to top